OpenAI has introduced a new feature called shopping research inside ChatGPT which can help users in getting detailed purchase recommendations. The users can access the tool on web and mobile for logged-in users across Free, Go, Plus and Pro plans, with OpenAI offering nearly unlimited usage through the holiday season.
In a blog post, the company stated that the feature aims to help users who rely on ChatGPT to evaluate products and compare options, particularly in categories where decisions require deeper context. All the user needs to do is instead of browsing multiple websites, users can describe what they need, such as a quiet vacuum cleaner, a budget gaming laptop or a gift for a family member and ChatGPT generates a customised buying guide.
The company stated that it works as a conversational research assistant. It asks follow-up questions to understand preferences, trusted retail sources for current information and narrows down choices based on specifications, pricing and availability. It can help well in specific categories that involve detailed comparisons including electronics, beauty, home, graden and more.
The shopping research is powered by GPT-5 mini, which has been trained specifically for product-focused tasks. It generates tailored recommendations using trusted web sources, real-time interaction, and personal context (when memory is enabled). Users can influence the search by marking products as “not interested” or requesting “more like this,” allowing the model to fine-tune its recommendations. The process concludes with a personalised buyer’s guide summarising the best options, key differences, and tradeoffs.
The feature is also integrated into ChatGPT Pulse for Pro users, allowing it to make proactive suggestions based on previous conversations, such as recommending accessories after a bike discussion.
Start by asking ChatGPT a shopping question like Best TVs to buy
When the shopping research option pops up, tap it to begin.
ChatGPT will ask a few quick questions about your budget, preferences, or who the item is for.
As it researches, you’ll see product suggestions come in.
Tell it what you like or don’t like by tapping Not interested or More like this.
After a few minutes, you’ll get a clean, personalised buyer’s guide with the best options and key differences.